JP2003140688A - System and method for information distribution - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information distribution system which can provide various distributed information without requiring any trouble to obtain and register the distributed information on an administrator side. SOLUTION: A voice information reception part 14 receives voice data sent from an information communication terminal through the Internet 200. A property extraction part 16 extracts properties of the voice data. A database registration part 16 registers the received voice data as the distributed information in a provided voice information storage part 20 while making it correspond to the properties. A database retrieval part 18 retrieves voice data corresponding to the properties supplied from the property extraction part 16 from pieces of voice data which are already stored in the provided voice information storage part 20. An information distribution part 19 sends the voice data supplied from the database retrieval part 18 to an information communication terminal through the Internet 200. The sent voice data are finally converted into a voice and outputted from the information communication terminal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information distribution system and an information distribution method, and more specifically, an information distribution system for distributing various distribution information from an information distribution server to a plurality of user terminals, and an information distribution system. The present invention relates to an information distribution method, for example, an information distribution system and an information distribution method that can be suitably used for an interactive pet robot or the like for promoting communication by an elderly person as a part of a welfare support system.

[0002]

2. Description of the Related Art Conventionally, an interactive pet type robot has been realized as a means for solving the problem that an elderly person who lives apart from his / her family alone and falls into communication shortage. When the user talks to the pet robot, the pet robot responds to the user according to the content of the conversation. Hereinafter, as a first conventional example,
This pet robot will be briefly described.

FIG. 23 shows the structure of a pet robot. The pet robot includes a microphone 900, a voice recognition unit 901, an analysis unit 902, a dialogue unit 903, a response generation unit 904, a dialogue database 905, a response unit 906, a voice data storage unit 907, and an output unit. And 908. The dialogue database 905 is an analysis dictionary 909.
And a dialogue dictionary 910 and a response dictionary 911. When the user talks to the pet robot, the voice is converted into voice data by the microphone 900. The voice data is voice-recognized by the voice recognition unit 901 and output as text data. The analysis unit 902 refers to the analysis dictionary 909 and converts the input character string, which is the text data, into an input concept. For example, to convert to the input concept which means "greetings" to the input character string "Hello" (tentatively referred to as "HELLO"). The dialogue unit 903 refers to the dialogue dictionary 910 and outputs an output concept corresponding to this input concept. For example,
For the input concept "HELLO" which means "greeting", the output concept (provisionally indicated as "\ HELLO") which also means "reply to greeting" is output. The response generation unit 904 refers to the response dictionary 911 and randomly selects one output character string from among the response character string candidates corresponding to the output concept, and outputs the selected output character string. For example, in response dictionary 911, as a response character string corresponding to the output concept "¥ HELLO", "Hi, Hello", "Hi, Hi", "Hello", "Hello Genki?"
If you but four had been registered, the response generation unit 904, one of the response string in this, for example, "Hi Genki?"
To output. In reality, the response generation unit 904 outputs the file name of the voice data corresponding to the selected response character string. In the voice data storage unit 907, the response dictionary 91
The voice data corresponding to the response character string registered in No. 1 is stored in advance, and the response unit 906 stores the voice data corresponding to the voice file name output from the response generation unit 904 from the voice data storage unit 907. Read out. The output unit 908 outputs a sound based on the read sound data.

[0004] With the above operation, for example, and say "Hello" users to the pet-type robot,
In response to this, pet-type robot, "Hi, Hello" to the user and "Hello, how are you" to reply and so on. Also, for example, when the user speaks "How are you?", He replies, "I'm fine. How about you?" In this way, the pet robot appropriately responds to the contents spoken by the user,
The user can enjoy the conversation as if he were talking to a real person.

On the other hand, in recent years, it has been considered that the pet type robot as described above has an information distribution function in order to use it for the purpose of nursing care, and such a system is currently realized. Specifically, a center device installed in a facility such as a care center distributes various information via a communication line to pet-type robots as information communication terminals owned by each user. An information distribution system has been realized in which information is notified to a user through a pet robot. Hereinafter, this information distribution system will be briefly described as a second conventional example.

FIG. 24 shows the overall configuration of the information distribution system. The information distribution system has a plurality (here, three) operated by a plurality of users (here, users A to C).
Information terminal devices 912a to 912c, the Internet 913, and a center device 914. The center device 914 and each of the information terminal devices 912a to 912c are communicably connected via the Internet 913.

FIG. 25 shows the configuration of the center device 914. The center device 914 has a distribution information database 91.
5, an information selection unit 916, and an information distribution unit 917. The distribution information database 915 includes the distribution information storage unit 9
18 and a voice data storage unit 919. The distribution information database 915 stores distribution information to be distributed to users. For example, the administrator of the center device 914 is
If there is distribution information to be transmitted to the users A to C, it is appropriately registered in the distribution information database 915. At this time, the response concept to which this delivery information should be output is also registered. For example, when the distribution information to be registered is information about today's schedule, it is registered that the distribution information corresponds to the response concept (tentatively indicated as “¥ YOTEI”). The distribution information may be voice data input by a manager from a microphone, text data input from a keyboard, or other data. The voice data is stored in the voice data storage unit 919, and the text data and other data are stored in the distribution information storage unit 918. The information selection unit 916 selects, for example, undistributed data newly added to the distribution information database 915 from the data registered in the distribution information database 915, and the information distribution unit 917 selects the selected data. Is distributed to the information communication terminals 912a to 912c.

FIG. 26 shows the configuration of the information communication terminal 912a. The information communication terminals 912a to 91 shown in FIG.
Since the configuration of 2c is the same, here, the information communication terminal 9
Only 12a will be described, and description of the information communication terminals 912b and 912c will be omitted. Also, FIG.
6, the same components as those shown in FIG. 23 are designated by the same reference numerals, and the description thereof will be omitted. The information communication terminal 912a further includes an information receiving unit 921 and an information registration unit 922 as compared with the configuration of the first conventional example shown in FIG. 23, and the dialogue database 920 has an information response dictionary 923.
Further includes. The information receiving unit 921 receives the distribution information distributed from the information distribution unit 917 of the center device 914 as described above. The information registration unit 922 analyzes the received information, and based on the received information, the information response dictionary 92.
3 to register the reception information and to register the voice data in the voice data storage unit 907. In this way, each time new distribution information is received, the registered content of the information response dictionary 923 is updated, and the received voice data is added to the voice data storage unit 907. In this state, for example, when the user speaks to the information communication terminal 912a, "What's your schedule for today?", This voice is supplied to the analysis unit 902 as text data through the microphone 900 or the like, and the analysis unit 902.
Outputs the input concept (tentatively labeled "YOTEI"),
The dialogue unit 903 outputs the response concept “¥ YOTEI” corresponding to the input concept “YOTEI”. Response generator 9
When 04 receives this response concept “¥ YOTEI”,
The response dictionary 91 for this response concept "\ YOTEI"
The information response dictionary 923 is referred to instead of 1, and the file name of the audio data corresponding to this response concept “¥ YOTEI” is output. The response unit 906 reads the voice data corresponding to this file name from the voice data storage unit 907 based on this file name. The output unit 908 outputs a voice based on this voice data. As a result, for example, in response to the user's question "What is today's schedule?", The information communication terminal 912a replies, "Today is the day of regular checkup."

As described above, by registering the information to be distributed to each user in the center device 914, this information can be notified to the user through the information communication terminal of the user. As a result, the user can expect a new topic, so that the freshness is not reduced even after long-term use. Further, since the distribution information is appropriately output in response to the user's inquiry, the user can obtain the information from the center device by only asking the information communication terminal, without requiring a special operation with a keyboard or the like. Therefore, even an elderly user who is not good at operating a keyboard or the like can easily obtain information.

[0010]

However, in the above-mentioned first conventional example, the pet robot only responds to the user's inquiry based on the response character string registered in the response dictionary in advance. For this reason, the dialogue tends to be monotonous, and if the pet robot is used for a long period of time, the freshness of the pet robot is reduced and the user becomes tired.

On the other hand, in the above-mentioned second conventional example, by the administrator of the center device registering the distribution information in the distribution information database, new information can be output from the information terminal device. However, since it takes a lot of work to obtain and register the distribution information, the number of distribution information that can be provided is limited.

When the same delivery information is delivered to all users, it is often useful information for some users but unnecessary for other users.
Not only is the storage capacity of the information terminal device wasted, but the convenience for the user is also reduced. In order to solve this problem, the center side may preliminarily specify the destination for each piece of distribution information, but the more the number of distribution destinations and distribution information, the more the work of specifying the destination. Takes time and effort.

Therefore, an object of the present invention is to provide an information distribution system capable of distributing more diverse information to an information communication terminal without requiring the administrator to obtain and register distribution information. It is to be.

Another object of the present invention is to provide useful distribution information to each user without requiring the administrator to specify the distribution destination of information for each distribution information. It is to provide a possible information distribution system.

[0015]

[Means for Solving the Problems and Effects of the Invention] A first invention is an information distribution system for distributing various distribution information from a center device to a plurality of information communication terminals through a communication line. Voice input means for converting the voice into voice data, attribute extracting means for extracting the attribute of the voice data from the voice data converted by the voice input means, and voice data converted by the voice input means by the attribute extracting means. A storage unit that stores the extracted attribute in association with it, a search unit that searches for voice data having a desired attribute from a plurality of voice data that is already stored by the storage unit, and a voice that is searched by the search unit. An information distribution means for distributing data to a predetermined information communication terminal, and converting the voice data distributed by the information distribution means into voice. A voice output means for inputting the voice, the voice input means and the voice output means are provided in each of the plurality of information communication terminals, and the storage means, the search means, and the information distribution means are provided in the center device. .

As described above, according to the first aspect of the invention, the voice uttered by the user is distributed to other users as distribution information, so that the administrator of the center device or the like can obtain the distribution information and register it. The distribution information is newly registered at any time without requiring the user, and the user can obtain fresh information. Further, since the voice data is stored in association with the attribute, it is easy to search and deliver the voice data having the desired attribute from the plurality of stored voice data. Further, since the attribute of the voice data is automatically extracted from the voice data itself, it is possible to save the trouble of the administrator or the like classifying the voice data according to the content.

According to a second aspect of the present invention, in the first aspect, the attribute extracting means includes a voice recognizing means for converting voice data into text data, a semantic analyzing means for extracting a meaningful keyword from the text data, and a keyword. And an attribute discriminating means for discriminating the attribute of the voice data based on the audio data.

As described above, according to the second invention, the attribute can be automatically extracted from the voice data itself.

In a third aspect based on the first aspect, the search means has an attribute corresponding to the attribute of the voice data converted by the voice input means from a plurality of voice data already stored by the storage means. Search voice data,
The information distribution means is characterized by distributing the voice data searched by the search means to the information communication terminal of the transmission source of the voice data converted by the voice input means.

As described above, according to the third aspect of the invention, since information from another user relating to a topic spoken by the user to the information communication terminal is output as voice of the other user through the information communication terminal, It is possible to obtain information in a form closer to a natural conversation, and there is no problem that information that is completely uninteresting to the user is output.
In addition, the user can simply talk to the information communication terminal about the topic of interest, and the information related to the topic is output as voice, so that the desired information can be obtained without the troublesome operation of a keyboard or the like. be able to.

In a fourth aspect based on the first aspect, the center device sets the attributes extracted by the attribute extraction means to
The search means further includes a preference attribute storage unit that stores the preference attribute of the user in association with the user, and the search unit refers to the preference attribute of the user stored by the preference attribute storage unit and is already stored by the storage unit. It is characterized in that the voice data having the attribute corresponding to the preference attribute of the user of the information delivery destination is searched from the plurality of voice data.

As described above, according to the fourth aspect of the invention, the voice data that matches the user's taste is searched and distributed, so that there is no problem that information that is completely uninteresting to the user is output. . Further, since unnecessary information is not transmitted to the information communication terminal, it is possible to reduce the traffic related to the information distribution and the storage capacity required for the information communication terminal when collectively transmitting the distribution information to the information communication terminal. Is reduced. In addition, it is possible for the administrator and the like to save the trouble of selecting the delivery destination and the delivery information in consideration of the user's preference. In addition, since the user's preference is automatically registered based on the content of the voice that the user has uttered in the past,
The user and the administrator can save the trouble of manually operating the keyboard or the like to manually register the preference attributes.

In a fifth aspect based on the fourth aspect, the attribute extracting means is provided in each of at least a plurality of information communication terminals, and the information distributing means converts the voice data searched by the searching means into the voice data. And a plurality of voice data stored by the received voice information storage means, in which the information communication terminal stores the voice data delivered by the information delivery means in association with the attribute. Further includes terminal-side search means for searching voice data of an attribute corresponding to the attribute extracted by the attribute extraction means, and the voice output means converts the voice data searched by the terminal-side search means into voice. And output.

As described above, according to the fifth aspect of the invention, in particular, even when the communication line is not connected, when the user speaks to the information communication terminal, the received voice information storage means stores it. From the stored voice data, voice data related to the topic spoken by the user is selected and output as voice from the information communication terminal. Therefore,
Even if the communication line is not connected, the user can obtain the information related to the content spoken appropriately in the interactive mode, so that the communication cost can be suppressed.

In a sixth aspect based on the fifth aspect, the information communication terminal temporarily stores the voice data converted by the voice input means while the communication line is not connected until the communication line is connected. Voice information temporary storage means for temporarily storing.

As described above, according to the sixth aspect of the present invention, the contents spoken by the user in the state where the communication line is not connected are the voice information temporary as voice data until the communication line is connected. The data is temporarily stored in the storage means and collectively transmitted to the center device when the communication line is connected. Therefore, it becomes possible to deliver the content spoken by the user to other users as the delivery information without wasting the content spoken by the user when the communication line is not connected.

A seventh invention is an information distribution system for distributing various distribution information from a center device to a plurality of information communication terminals through a communication line, and a voice input for converting a voice uttered by a user into voice data. Means, storage means for storing the voice data converted by the voice input means, information delivery means for delivering the voice data stored by the storage means to a predetermined information communication terminal, and voice delivered by the information delivery means. Speaker verification means for matching the speaker of the data to identify the speaker of this voice data, and speaker information storage means for storing priority information indicating the priority of the voice data of this speaker for each speaker. A voice output means for converting the delivered voice data into voice and outputting the voice. The voice input means, the speaker verification means, the speaker information storage means and the voice output means are provided with a plurality of information communication means. Provided in each of the terminal, storage means and information delivery means is provided in the center apparatus,
The delivered voice data is converted into voice by the voice output unit based on the priority information corresponding to the speaker of the voice data stored in the speaker information storage unit and output. To do.

As described above, according to the seventh aspect of the invention, even when the voice information of an unspecified number of users is delivered as the delivery information, the speaker is identified from the delivered voice data itself, The output of this voice data can be limited according to the priority set for each speaker. Therefore, for example, it is possible to limit the delivered voice data to be output first if the voice data is from a family and not to be output from a person who makes an aversion.

According to an eighth aspect of the invention, in the seventh aspect, the information communication terminal provides the user with respect to the speaker of the voice output by the voice output means, based on the voice data converted by the voice input means. And a speaker information registration unit for registering priority information for this speaker in the speaker information storage unit on the basis of the judgment result of the impression judgment unit.

As described above, according to the eighth aspect of the invention, the speaker information is automatically registered without the user registering the speaker information in advance, and the priority is optimally set.

A ninth aspect of the present invention is an information communication terminal for receiving various distribution information distributed from a center device through a communication line, and a voice input means for converting a voice uttered by a user into voice data, and a voice input. Attribute extracting means for extracting the attribute of this voice data from the voice data converted by the means, and voice for transmitting the voice data converted by the voice input means to the center device together with the attribute of this voice data extracted by the attribute extracting means. The information transmission means and the voice output means for converting the voice data distributed from the center device into voice and outputting the voice.

As described above, according to the ninth aspect of the invention, since the voice uttered by the user can be provided to the center device as the distribution information, the manager of the center device or the like obtains and registers the distribution information. The distribution information is newly registered at any time without requiring such trouble, and the user can obtain fresh information. Further, since the voice data is transmitted to the center device together with the attribute, it is easy for the center device to retrieve and deliver the voice data having the desired attribute from the plurality of stored voice data. Further, since the attribute of the voice data is automatically extracted from the voice data itself, it is possible to save the trouble of the administrator or the like classifying the voice data according to the content.

In a tenth aspect based on the ninth aspect, the attribute extraction means includes a voice recognition means for converting voice data into text data, a meaning analysis means for extracting a meaningful keyword from the text data, and a keyword recognition means. And an attribute discriminating means for discriminating the attribute of the voice data based on the audio data.

As described above, according to the tenth aspect, the attribute can be automatically extracted from the voice data itself.

In an eleventh aspect based on the ninth aspect, the voice data is distributed from the center device together with the attributes of the voice data, and the voice data distributed from the center device is stored in association with the attribute. The information storage means, and a terminal side search means for searching the voice data of the attribute corresponding to the attribute extracted by the attribute extraction means from the plurality of voice data stored by the received voice information storage means, The voice output means is characterized by converting the voice data searched by the terminal side search means into a voice and outputting the voice.

As described above, according to the eleventh invention, in particular, even when the communication line is not connected, when the user speaks to the information communication terminal, the received voice information case means is displayed. From the stored voice data, voice data related to the topic spoken by the user is selected and output as voice from the information communication terminal. Therefore,
Even if the communication line is not connected, the user can obtain the information related to the content spoken appropriately in the interactive mode, so that the communication cost can be suppressed.

The twelfth invention is the eleventh invention, wherein
It further comprises voice information temporary storage means for temporarily storing the voice data converted by the voice input means while the communication line is not connected until the communication line is connected.

As described above, according to the twelfth aspect of the invention, the contents spoken by the user in the state where the communication line is not connected are sequentially recorded as voice data as voice data until the communication line is connected. The data is temporarily stored in the storage means and collectively transmitted to the center device when the communication line is connected. Therefore, the content spoken by the user in a state where the communication line is not connected can be provided to the center device as distribution information without wasting.

A thirteenth invention is an information communication terminal for receiving various distribution information distributed from a center device through a communication line, and a voice input means for converting a voice uttered by a user into voice data, and the center device. A speaker collating means for collating the speaker of the voice data distributed from to identify the speaker of this voice data, and a talk for storing the priority information indicating the priority of the voice data of this speaker for each speaker. Person information storage means and voice output means for converting the delivered voice data to voice and outputting the voice, and the delivered voice data is stored in the speaker information storage means. It is characterized in that it is converted into a voice and outputted by the voice output means based on the priority information corresponding to.

As described above, according to the thirteenth invention, even when the voice information of an unspecified number of users is delivered as the delivery information, the speaker is identified from the delivered voice data itself, The output of this voice data can be limited according to the priority set for each speaker. Therefore,
For example, it is possible to limit the delivered voice data to be output first if the voice data is from a family and not to be output from a person who makes an aversion.

A fourteenth invention is the thirteenth invention, wherein
Based on the voice data converted by the voice input means, the favorable impression determination means for determining the impression the user has on the speaker of the voice output by the voice output means, and the determination result by the favorable impression determination means Based on the speaker, a speaker information registering unit for registering the priority information for the speaker in the speaker information storing unit is further provided.

As described above, according to the fourteenth invention, the speaker information is automatically registered without the user registering the speaker information in advance, and the priority is optimally set.

A fifteenth invention is a center device for distributing various distribution information to a plurality of information communication terminals through communication lines, and a voice information receiving means for receiving voice data from the information communication terminals. Storing means for storing the voice data associated with the voice data in association with the attribute of the voice data; searching means for searching voice data having a desired attribute from a plurality of voice data already stored by the storing means; And an information distribution unit that distributes the voice data searched by to a predetermined information communication terminal.

As described above, according to the fifteenth invention, the voice uttered by the user is distributed as distribution information to other users, so that the administrator of the center device or the like can obtain and register the distribution information. The distribution information can be newly registered at any time without needing, and fresh information can be distributed to the user. Further, since the voice data is stored in association with the attribute, it is easy to search and deliver the voice data having the desired attribute from the plurality of stored voice data.

The sixteenth invention is the fifteenth invention, wherein
The voice information receiving means further comprises attribute extracting means for extracting an attribute of the voice data from the voice data, and the storing means stores the received voice data in association with the attribute extracted by the attribute extracting means. It is characterized by

As described above, according to the sixteenth invention, since the attribute of the voice data is automatically extracted from the voice data itself, it is troublesome for the administrator or the like to classify the voice data according to the contents. I can omit it.

The 17th invention is the same as the 15th invention,
The attribute extraction unit includes a voice recognition unit that converts voice data into text data, a semantic analysis unit that extracts a meaningful keyword from the text data, and an attribute determination unit that determines the attribute of the voice data based on the keyword. .

As described above, according to the seventeenth invention, the attribute can be automatically extracted from the voice data itself.

The eighteenth invention is the same as the fifteenth invention,
The retrieval means retrieves the voice data having the attribute corresponding to the attribute of the voice data received by the voice information receiving means from the plurality of voice data already stored by the storage means, and the information distributing means receives the voice information reception. The voice data searched by the search means is distributed to the information communication terminal of the transmission source of the voice data received by the means.

As described above, according to the eighteenth aspect of the invention, the voice data transmitted from the information communication terminal and the information from another user related to the topic can be automatically distributed to the information communication terminal. There is no problem of delivering information that is completely uninteresting to the user.

The 19th invention is the same as the 15th invention,
The user further includes a preference attribute storage unit that stores the attribute of the voice data received by the voice information receiving unit in association with the user as the user's preference attribute, and the search unit is the user stored by the preference attribute storage unit. With reference to the preference attribute of, the voice data of the attribute corresponding to the preference attribute of the user of the information distribution destination is searched from the plurality of voice data already stored by the storage means.

As described above, according to the nineteenth invention, the voice data matching the user's taste is searched and distributed, so that there is no problem of distributing information that is not of interest to the user. . Moreover, since unnecessary information is not transmitted to the information communication terminal, it is possible to reduce traffic related to information distribution. In addition, it is possible for the administrator and the like to save the trouble of selecting the delivery destination and the delivery information in consideration of the user's preference. In addition, since the user's preference is automatically registered based on the content of the voice data received in the past from the information communication terminal, the user or the administrator purposely operates the keyboard or the like to manually register the preference attribute. You can save time and effort.

A twentieth aspect of the present invention is an information distribution method for distributing various distribution information from a center device to a plurality of information communication terminals through a communication line, and a voice input for converting a voice uttered by a user into voice data. Step, an attribute extraction step of extracting the attribute of this voice data from the voice data converted by the voice input step, and the voice data converted by the voice input step are stored in association with the attribute extracted by the attribute extraction step. The storing step, the searching step for searching the voice data having a desired attribute from the plurality of voice data already stored by the storing step, and the voice data searched by the searching step are distributed to a predetermined information communication terminal. And the audio data distributed by the information distribution step. And an audio output step of converting and outputting the voice input step and the audio output step is performed at each of a plurality of information communication terminals,
The storing step, the searching step and the information distributing step are characterized in that they are executed in the center device.

As described above, according to the twentieth aspect of the invention, since the voice uttered by the user is distributed to other users as distribution information, the administrator of the center device or the like can obtain the distribution information and register it. The distribution information is newly registered at any time without requiring the user, and the user can obtain fresh information. Further, since the voice data is stored in association with the attribute, it is easy to search and deliver the voice data having the desired attribute from the plurality of stored voice data. Further, since the attribute of the voice data is automatically extracted from the voice data itself, it is possible to save the trouble of the administrator or the like classifying the voice data according to the content.

The twenty-first invention is an information distribution method for distributing various distribution information from a center device to a plurality of information communication terminals through a communication line, and a voice input for converting a voice uttered by a user into voice data. Step, a storage step of storing the voice data converted by the voice input step, an information delivery step of delivering the voice data stored by the storage step to a predetermined information communication terminal, and a voice delivered by the information delivery step. A speaker verification step of matching the speaker of the data to identify the speaker of this voice data, and a speaker information storage step of storing priority information indicating the priority of the voice data of this speaker for each speaker. , A voice output step of converting the delivered voice data into voice and outputting the voice, the voice input step, the speaker verification step, the speaker information The storage step and the voice output step are executed in each of the plurality of information communication terminals, the storage step and the information distribution step are executed in the center device, and the distributed voice data is stored in the speaker information storage step. Based on the priority information corresponding to the speaker of the voice data, the voice output step converts the voice into voice and outputs the voice.

As described above, according to the twenty-first aspect, even when the voice information of an unspecified number of users is delivered as the delivery information, the speaker is identified from the delivered voice data itself, The output of this voice data can be limited according to the priority set for each speaker. Therefore,
For example, it is possible to limit the delivered voice data to be output first if the voice data is from a family and not to be output from a person who makes an aversion.

[0057]

BEST MODE FOR CARRYING OUT THE INVENTION Various embodiments of the present invention will be described below with reference to the drawings. (First Embodiment) FIG. 1 shows the overall configuration of an information distribution system according to a first embodiment of the present invention. The information distribution system includes three information communication terminals 100a to 100c operated by users A to C and the Internet 2 respectively.
00 and a center device 300. Each of the information communication terminals 100a to 100c is connected to the center device 300 via the Internet 200.

FIG. 2 shows the configuration of the information communication terminal 100a. Since the information communication terminals 100a to 100c have the same configuration, only the configuration of the information communication terminal 100a will be described here and the description of the information communication terminals 100b and 100c will be omitted. As shown in FIG. 2, the information communication terminal 100a.
Includes a microphone 10, a voice information transmitting unit 11, an information receiving unit 12, and a voice output unit 13. The user (user A in this case) speaks to the information communication terminal 100a. The microphone 10 converts the voice uttered by this user into voice data. The voice information transmitter 11 sends this voice data to the center device 3 via the Internet 200.
To 00. However, in reality, the user ID and the like are transmitted together with the voice data. The information receiving unit 12 receives voice data from the center device 300 via the Internet. The voice output unit 13 outputs the received voice data as voice.

FIG. 3 shows the configuration of the center device 300. The center device 300 includes a voice information reception unit 14, a database registration unit 15, an attribute extraction unit 16, a distribution information database 17, a database search unit 18, and an information distribution unit 19. The distribution information database 17 has a provided voice information storage unit 20. Voice information receiving unit 14
Receives the voice data transmitted from the information communication terminals 100a to 100c. The received voice data is supplied to the database registration unit 15 and the attribute extraction unit 16. The attribute extraction unit 16 extracts an attribute of voice data. Here, the "attribute" indicates the classification of the content of the voice data, that is, the content of the story spoken by the user toward the information communication terminal. For example, if the content spoken by the user is about fishing, the attribute extraction unit 16 extracts "fishing" as the attribute of this voice data. FIG. 4 shows a specific configuration of the attribute extraction unit 16. The attribute extraction unit 16 includes the voice recognition unit 2
1, a semantic analysis unit 22, and an attribute determination unit 23.
The voice recognition unit 21 voice-recognizes voice data and outputs text data. The meaning analysis unit 22 analyzes the meaning of this text data and outputs meaningful words as keywords. The attribute discrimination unit 23 discriminates the attribute from the keyword. For example, when the voice recognition unit 21 outputs the text “I went fishing today” as text data, the semantic analysis unit 22 outputs the keywords “Today” and “Fishing” from the text data. The attribute determination unit 23 determines that the “attribute” of this voice data is “fishing” based on these keywords. Returning to FIG. 3, the attributes extracted by the attribute extraction unit 16 are supplied to the database registration unit 15 and the database search unit 18. The database registration unit 15 registers the received voice data as delivery information in the provided voice information storage unit 20 in the delivery information database 17. At this time, as shown in FIG. 5, the voice data is stored in the provided voice information storage section 20 in association with the attribute. Here, for example, the new voice data V06 is newly stored in association with the attribute “fishing”. In the provided voice information storage unit 20, the users A to C transmitted from the information communication terminals 100a to 100c in this manner.
Voice data is stored together with attributes.

The database retrieval unit 18 retrieves the voice data corresponding to the attribute supplied from the attribute extraction unit 16 from the voice data stored in the provided voice information storage unit 20, and retrieves it to the information distribution unit 19. The audio data that has been processed. Here, for example, based on the attribute “fishing” extracted by the attribute extracting unit 16, voice data having the attribute “fishing” is searched, and as a result, the voice data V01 to voice data V04 are selected and the information is displayed. It is supplied to the distribution unit 19. Which of the voice data V01 and the voice data V04 is selected depends on the provided voice information storage unit 2.
The older timing stored in 0 may be preferentially selected or may be selected randomly. Information distribution unit 19
Uses the voice data supplied from the database search unit 18 via the Internet 200 to the information communication terminal 100a.
Send to. At this time, the reply destination address stores the correspondence between the user ID and the reply destination address, for example, based on the user ID notified together when the voice data is transmitted from the information communication terminal 100a. You can find out by referring to the database.
The voice data transmitted by the information distribution unit 19 is finally output as voice from the voice output unit 13 of the information communication terminal 100a as described above.

The operation of the center device 300 will be described with reference to the flowchart shown in FIG. In step S10, the center device 300 receives data including voice data. In step S11, the received data is analyzed and expanded. In step S12, the attribute of the received voice data is extracted. In step S13, the voice data is registered in the distribution information database 17 based on the extracted attributes. In step S14, based on the extracted attributes,
The distribution information database 17 is searched for audio data corresponding to this attribute. In step S15, the voice data selected by the search is transmitted.

In the first embodiment, by the above operation,
For example, when the user A speaks something to the information communication terminal 100a, one voice data related to the topic spoken by the user A is selected from the voice data of other users already registered in the center apparatus 300. (For example,
The voice data registered by the user B talking to the information communication terminal 100b before) is selected and output as voice from the information communication terminal 100a.
Further, the content spoken by the user A is registered in the center device 300 as voice data, and the other users B and C relate to the content spoken by the user A to the information communication terminals 100b and 100c, respectively. When talking about a topic, it is transmitted to the information communication terminals 100b and 100c and output as voice.

More specifically, for example, when the user A talks to the information communication terminal 100a, "I went fishing today," the center apparatus 300 has already been reached.
From the voice information registered in, the voice information “I can catch such a fish in the lake XX now” is selected as the voice information related to the topic spoken by the user A, and the information communication terminal 100a of the user A is selected. Is output by voice. At this time,
The voice information "I went fishing today" spoken by the user A is registered in the center device 300. After that, for example, when the user B talks to the information communication terminal 100b about "fishing", The voice information of the user A is selected, and is output from the information communication terminal 100b of the user B by voice, "I went fishing today."

As described above, according to the first embodiment, the contents spoken by the user to the information communication terminal are directly registered in the center device as distribution information and distributed to other users. This saves the administrator the trouble of obtaining the distribution information and registering it in the distribution information database, and the management cost of the information distribution system can be reduced. Further, since the voice data is automatically registered in the database in association with the attribute by the attribute extraction unit,
The administrator does not have to manually determine the attribute of voice data and register it. In addition, since the distribution information is newly registered at any time, the user can obtain fresh information. In addition, since information from another user related to a topic spoken to the information communication terminal by the user is output by the voice of the other user through the information communication terminal, it is possible to obtain information in a form closer to a natural conversation. It is possible, and there is no problem that information that is completely uninteresting to the user is output.
In addition, the user can simply talk to the information communication terminal about the topic of interest, and the information related to the topic is output as voice, so that the desired information can be obtained without the troublesome operation of a keyboard or the like. be able to. In addition, the administrator can save the trouble of designating the delivery destination for each delivery information.

In the first embodiment, the attribute extraction unit 1
Although 6 is provided in the center device 300, the present invention is not limited to this. For example, the attribute extraction unit 16 is set to each information communication terminal 10
You may provide in 0a-100c. Specifically, FIG.
The attribute extraction unit 16 is provided in the subsequent stage of the microphone 10, the attribute of the voice data output from the microphone 10 is extracted by the attribute extraction unit 16, and the extracted attribute is transmitted to the center device 300 together with the voice data. The center device 300 registers the voice data in the database based on the received attribute and searches the database based on the received attribute.

Furthermore, a part of the configuration of the attribute extraction unit 16 may be provided in the information communication terminals 100a to 100c, and the other part may be provided in the center device 300.
7 and 8 show the configurations of the information communication terminal 101a and the center device 301 as a modified example. 7 or 8, the same components as those shown in FIGS. 2 to 4 are designated by the same reference numerals. 7, in addition to the configuration of the information communication terminal 100a shown in FIG. 2, the information communication terminal 101a further includes a voice recognition unit 21 and a semantic analysis unit 21 which are part of the configuration of the attribute extraction unit 16 shown in FIG. Including 22. The voice data output from the microphone 10 is converted into keywords through the voice recognition unit 21 and the semantic analysis unit 22. This keyword is transmitted to the center device 301 together with the voice data. On the other hand, in FIG.
In the center apparatus 300 shown in FIG. 3, the attribute discriminating unit 23, which is a part of the constitution of the attribute extracting unit 16 shown in FIG. Voice information receiver 1
4 supplies the received voice data to the database registration unit 15 and also supplies the received keyword to the attribute discrimination unit 23. The attribute discriminating unit 23 discriminates the attribute from the supplied keyword. Subsequent processing is the same as the configuration shown in FIG.

As described above, the attribute extraction unit 16 can be arbitrarily provided on the information communication terminal side or the center device side. By providing the attribute extraction unit 16 on the side of the center device, there is an advantage that the configuration of each information communication terminal is simplified. On the other hand, when the present invention is applied to an apparatus that originally has a voice recognition function, such as the pet robot shown in FIG. 23 as the first conventional example, the voice recognition unit 2
It is preferable to provide the configuration of the attribute extracting unit 16, particularly the voice recognizing unit 21, on the information communication terminal side in that the configuration can be simplified by sharing the configuration of 1.

(Second Embodiment) In the first embodiment,
Every time the user talks to the information communication terminal, it is necessary to perform a process of searching the distribution information database in the center device and transmitting voice data of a common topic to the information communication terminal. Therefore, the user must keep the communication line connected while interacting with the information communication terminal. Therefore, as a second embodiment,
A distribution information system capable of shortening the connection time of the communication line while realizing the same function as that of the first embodiment will be described.

The overall system according to the second embodiment of the present invention is the same as that of the first embodiment shown in FIG. However, the information communication terminal 100a-
Instead of 100c and the center device 300, information communication terminals 102a to 102c and a center device 302 described later are provided.

FIG. 9 shows the configuration of the information communication terminal 102a. The configuration of the information communication terminals 102b and 102c is as follows.
Since the configuration is the same as that of the information communication terminal 102a, the description will be omitted. The information communication terminal 102a includes the microphone 10
An attribute extraction unit 16, a voice information temporary storage unit 24, a voice information transmission unit 11, an information reception unit 12, a database registration unit 25, a received voice information storage unit 26, a database search unit 27, and a voice. And an output unit 13. In FIG. 9, the same components as those shown in FIGS. 2 and 3 are designated by the same reference numerals, and the description thereof will be omitted.

The attribute extraction unit 16 extracts attributes from the voice data output from the microphone 10 and supplies them to the voice information temporary storage unit 24 and the database search unit 27.
The voice information temporary storage unit 24 temporarily stores the voice data output from the microphone 10 and the attribute extracted by the attribute extraction unit 16. These temporarily stored data are transmitted to the center device 302 together with the user ID through the voice information transmitting unit 11 at the timing when the communication line is properly connected. The processing in the center device 302 regarding these data will be described later. The information receiving unit 12 collectively receives a plurality of voice data from the center device 302 at the timing when the communication line is properly connected. The plurality of pieces of voice data are a set of voice data that matches the preference of the user (here, user A). However, in reality, information on the attribute of each audio data is also included. The database registration unit 25 stores the plurality of received voice data in the received voice information storage unit 26 in association with the attributes. Based on the attribute extracted by the attribute extraction unit 16, the database retrieval unit 27 retrieves the voice data related to this attribute from the plurality of voice data stored in the received voice information storage unit 26, and retrieves the voice data. Based on the result, one audio data is supplied to the audio output unit 13. The voice output unit 13 converts this voice data into voice and outputs it.

FIG. 10 shows the configuration of the center device 302. The center device 302 includes a voice information receiving unit 14, a database registration unit 15, and a distribution information database 17
, Preference attribute storage unit 28, and database search unit 18
And an information distribution unit 19. The distribution information database 17 includes a provided voice information storage unit 20. Note that FIG.
3, the same components as those in FIG. 3 are designated by the same reference numerals and the description thereof will be omitted.

The voice information receiving section 14 is connected to the information communication terminals 102a to 102 at the timing when the communication line is properly connected.
Receive the data sent from c. It should be noted that, as described above, this received data includes a plurality of audio data and information regarding attributes of the plurality of audio data,
The database registration unit 15 sequentially registers the voice data in the distribution information database 17 in association with the attribute based on the received data. On the other hand, the attribute information included in the received data is also supplied to the preference attribute storage unit 28. As shown in FIG. 11, the preference attribute storage unit 28 indicates the attribute as the user I of the transmission source user of the received data (here, user A).
It is associated with D (denoted as abc in the figure) and stored as a “preference attribute” which is a field that matches the preference of the user A.
This is because the topic in which the user A talks to the information communication terminal 102a is, in most cases, a topic that matches the taste of the user.
Every time a new attribute is supplied to a certain user ID, the preference attribute storage unit 28 adds the new attribute to the preference attribute of this user. For example, if the attribute “music” is newly supplied as the attribute of the voice data from the user A in the state shown in FIG. 11, the user ID
"Fishing" and "golf" as preference attributes for "abc"
Three attributes of "music" will be registered. The database search unit 18 searches the voice data stored in the delivery information database 17 for voice data that matches the taste of the destination user at the timing when the communication line is appropriately connected, and informs the information delivery unit 19 of the voice data. Supply. For example, if the destination user is user A, the database search unit 18 causes the preference attribute storage unit 28 to associate the preference attribute (here, “fishing” “golf” with the user ID “abc” of the user A). )), The voice data stored in the provided voice information storage unit 20 is searched for voice data corresponding to the preference attributes “fishing” and “golf”. For example, if the provided voice information storage unit 20 stores data as shown in FIG. 5, voice data V01 and voice data V03 to V05 are used as voice data corresponding to the preference attributes “fishing” and “golf”. elect. The plurality of selected voice data, together with their attributes, are transmitted through the information distribution unit 19 to the information communication terminal 1 of the user A.
02a. As a result, the voice data as shown in FIG. 12 is stored in the reception voice information storage unit 26 shown in FIG. The timing for connecting the communication line may be, for example, at regular intervals (every 10 minutes, for example), when the distribution information database 17 is updated, or at other timings. May be.

Here, referring again to FIG. 9, when the communication line is not connected, the user A uses the information communication terminal 102.
The operation when talking to a will be particularly described. The voice of the user A speaking to the information communication terminal 102a is converted into voice data by the microphone 10,
The attribute extraction unit 16 extracts an attribute from this voice data. Since the communication line is not connected for voice data,
It cannot be transmitted to the center device 302, and is temporarily stored in the voice information temporary storage unit 24. Thus, the voice data is sequentially stored in the voice information temporary storage unit 24 until the communication line is connected. Attribute extraction unit 16
The attribute extracted by is supplied to the database search unit 27. For example, when the user A talks to the information communication terminal 102a about fishing, the database search unit 27 is supplied with the attribute “fishing”. The database search unit 27 searches the voice data stored in the received voice information storage unit 26 for voice data related to the attribute extracted by the attribute extraction unit 16, and returns 1 as appropriate.
One audio data is selected and supplied to the audio output unit 13.
Specifically, from the voice data shown in FIG. 12, voice data (voice data V01, V
04), and one voice data is appropriately selected from the search results. As described above, this selection method may be determined by the timing of storage, may be random, or may be determined by another method. As a result, the voice output unit 13
Thus, the topic related to fishing that the other user has spoken is output by voice.

The operation of the information communication terminal 102a when the communication line is not connected will be described below with reference to the flowchart shown in FIG. Information communication terminal 102
In a, in step S20, the voice uttered by the user is converted into voice data. In step S21, the attribute of this voice data is extracted. In step S22, the voice data and its attribute are stored in the voice information temporary storage unit 24. In step S23, the voice data corresponding to the attribute extracted in step S21 is selected from the voice data stored in the received voice information storage unit 26 by searching. In step S24, the selected voice data is converted into voice and output.

Next, the operation of the center device 302 when receiving voice data will be described with reference to the flowchart shown in FIG. However, in FIG.
Steps that are the same as are labeled with the same reference numerals. In step S10, the center device 302 receives voice data (including the voice data attribute) from the information communication terminal.
In step S11, the received voice data is analyzed and expanded. In step S13, the provided voice information storage unit 20
The received attribute and voice data are registered in association with. In step S30, the received attribute is registered in the preference attribute storage unit 28 as the user's preference attribute.

Next, with reference to the flow chart shown in FIG. 15, the operation of the center device 302 at the time of voice data distribution will be described. However, in FIG.
Steps that are the same as are labeled with the same reference numerals. In the center device 302, in step S40, the preference attribute storage unit 2
Based on the user's preference attribute stored in 8, the voice data stored in the provided voice information storage unit 20 is searched for voice data that matches the user's preference. In step S15, the voice data selected by the search is transmitted.

As described above, according to the second embodiment,
In particular, even if the communication line is not connected, when the user speaks to the information communication terminals 102a to 102c, the user selects from the voice data stored in the received voice information storage unit 26. One voice data related to the talked topic is selected, and the information communication terminal 102 is selected.
Audio is output from a to 102c. Therefore, it is not necessary to always connect the communication line, and the communication cost can be suppressed. Further, the contents spoken by the user in the state where the communication line is not connected are sequentially stored as voice data in the voice information temporary storage unit 24 until the communication line is connected, and the communication line is connected. When it is performed, the information is collectively transmitted to the center device 302. Therefore, the content spoken by the user in a state where the communication line is not connected can be delivered to other users as information without wasting it. Further, since the voice data collectively transmitted from the center device 302 to the information communication terminals 102a to 102c are filtered and transmitted according to the preference of the user, communication traffic can be reduced, and the information communication terminals 102a to 102c can be reduced. It is possible to reduce the storage capacity of the device, and it is possible to avoid a problem caused by information unnecessary for the user being output from the information communication terminals 102a to 102c at some time. In addition, since the user's preference can be automatically extracted based on the contents spoken by the user, the user's preference can be manually registered by the administrator, or the user can register the user's preference using a keyboard or the like. You can save the trouble of registration.

(Third Embodiment) In a system for distributing a voice from a user as distribution information, as described above, how to transmit only necessary distribution information for each user and unnecessary distribution. The problem is whether to prevent information from being distributed, but in the information distribution system according to the first and second embodiments described above, the voice data from the user is classified according to the attribute and required by the user. This problem is solved by delivering only the audio data of the content. That is, it is determined whether or not to deliver the information to the user based on the content of the information, so that unnecessary information is not delivered to the user. On the other hand, in the third embodiment described below, the voice output from the information terminal device is restricted by determining on the information communication terminal side who is the provider of the voice information.

The overall system according to the third embodiment of the present invention is the same as that of the first embodiment shown in FIG. However, instead of the center device 300, the center device 302 shown in FIG.
Instead of 00a to 100c, an information communication terminal 1 described later
03a to 103c.

The information communication terminal 103a will be described below. Since the information communication terminals 103a to 103c have the same configuration, the description of the information communication terminals 103b and 103c will be omitted. Regarding the center device 302,
Since it has already been described in the second embodiment, the description is omitted here.

FIG. 16 shows the configuration of the information communication terminal 103a. The information communication terminal 103a includes the microphone 10 and
The attribute extraction unit 16, the voice information temporary storage unit 24, the voice information transmission unit 11, the information reception unit 12, and the speaker verification unit 29.
A speaker information search unit 30, a speaker information storage unit 31, a database registration unit 25, a received voice information storage unit 32,
The database search unit 27 and the voice output unit 13 are provided. In FIG. 16, the same components as those in FIG. 9 are designated by the same reference numerals, and the description thereof will be omitted.

The information receiving unit 12 receives the voice data (actually also including attributes) transmitted as distribution information from the center device 302. The received voice data is supplied to the speaker verification unit 29 and the database registration unit 25. The speaker verification unit 29 verifies who the speaker of the received voice data is among the pre-registered speakers by speaker verification, and the speaker of the received voice data is talked based on the verification result. The speaker ID of the speaker is output. The speaker ID is an identifier used to identify a speaker in the information communication terminal 103a, and is distinguished from a user ID used to identify a user in the center device 302. The speaker ID can be uniquely set in the information communication terminal 103a. The speaker information retrieval unit 30 refers to the speaker information storage unit 31 based on the speaker ID output from the speaker verification unit 29, and acquires the priority information corresponding to this speaker ID. FIG. 17 shows an example of data stored in the speaker information storage unit. The priority information represents the priority when the audio data is output, and here, as an example, the priority is 0-4.
The voice data of this speaker is never output when the priority information is set to "0". If the priority information is "4",
The voice data of this speaker is always output first. When the priority information is "1" and "3", the probability of output is 0.5 times and twice that of the case where the priority information is "2". The speaker information storage unit 31 can be set in advance by the user. For example, the priority information is set to “4” for the speaker IDs of family members so that the speaker information is always output first. It is conceivable to set the priority information to "0" for the speaker ID of the person who dislikes it so that it is not output.

The speaker information retrieval unit 30 notifies the database registration unit 25 of the acquired priority information. The database registration unit 25 registers the voice data supplied from the information reception unit 12 in the received voice information storage unit 32 in association with the attribute information and the priority information. At this time, it is not always necessary to store the voice data for the priority information “0”, but in the following description, it is assumed that the priority information “0” is also registered. FIG. 18 shows an example of data stored in the received voice information storage unit 32.
In this way, the received voice information storage unit 32 stores a plurality of voice data distributed from the center device 302 in association with the priority information.

In this state, the user A selects the information communication terminal 10
When talking about fishing, for example, with respect to 3a, the attribute "fishing" is extracted from this voice through the microphone 10 and the attribute extraction unit 16, and is notified to the database search unit 27. In response to this, the database search unit 27 searches for voice data regarding “fishing” from the plurality of voice data stored in the received voice information storage unit 32. At this time, the received voice information storage unit 32 is shown in FIG.
Assuming that the voice data shown in 8 is stored,
The voice data regarding "fishing" is voice data V01 and V04. However, since the priority information of the voice data V04 is set to "0", that is, the information not to be output, the database search unit 27 supplies the voice data V01 to the voice output unit 13, and the voice output unit 13 This voice data V01 is converted into voice and output. In the search process in the database search unit 27, for example, when two voice data whose priority information is “1” and “3” are searched as candidates, whichever is based on the probability calculated from the priority information. One of the voice data is selected. In this case, the voice data whose priority information is "3" is different from the voice data whose priority information is "1".
The probability of being elected is four times.

Next, the operation of registering the voice data received from the center device 302 in the received voice information storage unit 32 will be described with reference to the flowchart shown in FIG. The information communication terminal 103a receives the voice data from the center apparatus 302 in step S50. In step S51, speaker verification processing is performed based on the received voice data, and a speaker ID corresponding to this voice data is obtained. In step 52, the speaker information stored in advance in the speaker information storage unit 31 is referred to acquire the priority information set for the speaker ID obtained by the speaker verification.
In step S53, the received voice data is stored in the received voice information storage unit 32 in association with the priority information and the attribute.

As described above, according to the third embodiment,
It becomes possible for the user to arbitrarily set the output priority of the audio data according to the speaker. Therefore, for example, it is possible to preferentially output the voice information from the family and not to output the voice information from an unpolite person who often suffers slander or the like. Further, as an extreme example, a method of using such as not outputting voice information from anyone other than an acquaintance can be considered.

The present embodiment and the above-mentioned first and second
In the system according to the embodiment of the present invention, the voice information used as the distribution information is not only the information input by the user from the microphone 10 of the information communication terminal described above, but also does not have the user ID through other than that, for example, a general telephone. It is also possible to use the information input by an unspecified person. This has the advantage of increasing the number of sources of voice information. In the third embodiment, even if the voice data is from a person who does not have such a user ID, the speaker can be specified from the voice data itself if the voice data is registered in advance. You can limit the output.

(Fourth Embodiment) In the third embodiment, the output of voice data is limited based on the speaker information stored in the speaker information storage unit 31 in advance. In the system according to the fourth embodiment, it is possible to automatically register or update the speaker information appropriately based on the user's response content to the voice actually output from the information terminal device. The information delivery system according to the fourth embodiment will be described below. The fourth embodiment is different from the third embodiment in that the information communication terminal 10
Information communication terminal 104 described later instead of 3a to 103c
Since only a to 104c are provided, the description of the other same configurations will be omitted.

FIG. 20 shows the configuration of the information communication terminal 104a. Since the information communication terminals 104a to 104c have the same configuration, the description of the information communication terminals 104b and 104c will be omitted. The information communication terminal 104a includes a microphone 10, a voice recognition unit 21, a semantic analysis unit 22, an attribute determination unit 23, a voice information temporary storage unit 24, a voice information transmission unit 11, an information reception unit 12, and a talk. Speaker verification unit 29, speaker information retrieval unit 30, speaker information storage unit 31, database registration unit 25, received voice information storage unit 33, database retrieval unit 27, voice output unit 13, and favorable sensitivity. Judgment unit 3
4 and a speaker information registration unit 35. 20, the same components as those in FIG. 16 are designated by the same reference numerals, and the description thereof will be omitted.

The priority information of the voice data transmitted from the center device 302 is acquired from the speaker information storage unit 31 through the speaker verification unit 29 and the speaker information search unit 30. The speaker information retrieval unit 30 uses the priority information as the speaker ID.
Along with this, the database registration unit 25 is notified. The database registration unit 25 uses the received voice data as its attribute,
It is registered in the received voice information storage unit 33 in association with the priority information and the speaker ID. FIG. 21 shows the received voice information storage unit 3
3 shows an example of data stored in No. 3.

In this state, the user A selects the information communication terminal 10
3a, for example, when talking about fishing, after this voice is converted into voice data by the microphone 10, the voice recognition unit 21, the semantic analysis unit 22, and the attribute determination unit 23 (these configurations are shown in FIG. The attribute “fishing” is extracted through the attribute extraction unit 16) and is notified to the database search unit 27. Upon receiving this attribute, the database search unit 27 searches the voice data regarding “fishing” from the plurality of voice data stored in the received voice information storage unit 33. At this time, assuming that the voice data shown in FIG. 21 is stored in the received voice information storage unit 33, the voice data regarding “fishing” is the voice data V01 and V04. However, since the priority information of the voice data V04 is set to "0", that is, the information not to be output, the database search unit 27 supplies the voice data V01 to the voice output unit 13, and the voice output unit 13 This voice data V01 is converted into voice and output. At this time, the database search unit 27
Is the selected voice data (here, voice data V01)
The speaker ID is notified to the speaker information registration unit 35.

It is assumed that the user A hears the voice output from the voice output unit 13 as described above and speaks, for example, "I don't like this person very much". Then, the voice of the user A is converted into voice data by the microphone 10, converted into text data by the voice recognition unit 21, a keyword is extracted by the semantic analysis unit 22, and the keyword is supplied to the favorable determination unit 34. It The positivity determination unit 34 determines, based on the keyword, what impression the user A has on the speaker of the voice output immediately before from the voice output unit 13. Here, it is determined that the user A does not have a very good impression on the speaker of the output voice, and the determination result is notified to the speaker information registration unit 35. The speaker information registration unit 35 registers the speaker information in the speaker information storage unit 31 based on the speaker ID notified from the database search unit 27 and the determination result notified from the positivity determination unit 34. Here, since it can be seen that the speaker having the speaker ID “P001” does not have a good impression, the speaker information shown in FIG. 17, for example, is already registered in the speaker information storage unit 31. If so, the priority information for the speaker ID is changed from "2" to "1". Thus, the speaker information is automatically and optimally set without the need for the user to manually set it.

Next, the operation when the voice data from the speaker not yet registered in the speaker information storage section 31 is delivered will be described. The speaker verification unit 29 is the information receiving unit 1
As a result of performing the speaker verification process based on the voice data received by 2, when it is determined that the speaker of this voice data is not registered, this voice data is registered for speaker verification and the voice data is also registered. Speaker I suitable for the speaker
Allocate D and notify the speaker information search unit 30. The speaker information search unit 30 searches the speaker information corresponding to the speaker ID notified from the speaker verification unit 29, but the speaker information corresponding to this speaker ID is still the speaker information storage unit 31. Since it is not registered in the database, the priority information and the speaker ID are notified to the database registration unit 25 for the time being, assuming that the priority information is “2”. The database registration unit 25
The voice data is registered in the received voice information storage unit 33 in association with the attribute, the priority information and the speaker ID. In this state, it is assumed that, as a result of the user A speaking something, the voice data of the unregistered speaker is converted into voice by the voice output unit 13 and output. When the user A, who has heard the voice information from the unregistered speaker, makes some utterance in response to the voice information, the voice is converted into voice data by the microphone 10, and further, the voice recognition unit 21, the semantic analysis unit 22, and the favorable analysis unit. Through the sensitivity determination unit 34, the impression that the user A has for the unregistered speaker is determined based on the content of the statement made by the user A. The speaker information registration unit 35 newly adds the speaker information for this speaker ID to the speaker information storage unit 31 based on the speaker ID from the database search unit 27 and the determination result from the positivity determination unit 34. register. Accordingly, when the voice data of the speaker corresponding to this speaker ID is delivered from the next time onward, the priority is determined based on this speaker information.

Next, with reference to the flowchart shown in FIG. 22, the operation from the output of voice to the registration of speaker information in the information communication terminal 104a will be described.
Note that, in FIG. 22, the same processing steps as those in FIG. 13 are designated by the same step numbers, and detailed description thereof will be omitted. In step S60, the voice data stored in the received voice information storage unit 33 is searched for voice data corresponding to the attribute determined by the attribute determination unit 23. In step S24, the retrieved voice data is output as voice. In step S20, the voice output by the user is input to the output voice and converted into voice data. Step S6
In step 1, the voice data is converted into text data by the voice recognition process, and in step S62, a meaningful keyword is extracted from the text data. In step S63, the user's liking for the speaker is determined based on the extracted keywords. In step S64, the speaker information is newly registered or updated based on the result of the determination of the likeability.

As described above, according to the fourth embodiment,
The speaker information is automatically registered without the user registering the speaker information in advance, and the priority is optimally set. Therefore,
It is particularly effective in an information distribution system in which voice data from an unspecified number of speakers is distributed.

In the third and fourth embodiments, the output of voice data by speaker verification is limited based on the system of the second embodiment, but the present invention is not limited to this, and the output from the user is limited. It can be applied to all systems that distribute the audio data of. For example, as an example, all the voice data provided by the user through the information communication terminal or the telephone is distributed to each information communication terminal regardless of the content, and the information communication terminal selects from among the plurality of distributed voice data. It can also be applied to a system in which one audio data item is selected and output at an arbitrary timing.

[Brief description of drawings]

FIG. 1 is a block diagram showing an overall configuration of an information distribution system according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of an information communication terminal.

FIG. 3 is a block diagram showing a configuration of a center device.

FIG. 4 is a block diagram showing a configuration of an attribute extraction unit.

FIG. 5 is a diagram showing an example of data stored in a provided voice information storage unit.

FIG. 6 is a flowchart showing an operation of the center device.

FIG. 7 is a block diagram showing a configuration of an information communication terminal according to a modified example of the first exemplary embodiment.

FIG. 8 is a block diagram showing a configuration of a center device according to a modification of the first embodiment.

FIG. 9 is a block diagram showing a configuration of an information communication terminal in the information distribution device according to the second embodiment of the present invention.

FIG. 10 is a block diagram showing a configuration of a center device according to the second embodiment.

FIG. 11 is a diagram illustrating an example of data stored in a preference attribute storage unit.

FIG. 12 is a diagram showing an example of data stored in a received voice information storage unit.

FIG. 13 is a flowchart showing an operation of the information communication terminal when the communication line is not connected.

FIG. 14 is a flowchart showing an operation of the center device when receiving voice data.

FIG. 15 is a flowchart showing an operation of the center device at the time of voice data distribution.

FIG. 16 is a block diagram showing a configuration of an information communication terminal in an information distribution device according to a third embodiment of the present invention.

FIG. 17 is a diagram showing an example of data stored in a speaker information storage unit.

FIG. 18 is a diagram showing an example of data stored in a received voice information storage unit.

FIG. 19 is a flowchart showing an operation of registering voice data received from the center device in the information communication terminal.

FIG. 20 is a block diagram showing a configuration of an information communication terminal in an information distribution device according to a fourth exemplary embodiment of the present invention.

FIG. 21 is a diagram showing an example of data stored in a voice information storage unit.

FIG. 22 is a flow chart showing an operation from outputting a voice to registering speaker information in the information communication terminal.

FIG. 23 is a block diagram showing a configuration of a pet robot according to a first conventional example.

FIG. 24 is a block diagram showing an overall configuration of an information distribution system according to a second conventional example.

FIG. 25 is a block diagram showing a configuration of a center device in a second conventional example.

FIG. 26 is a block diagram showing a configuration of an information communication terminal in a second conventional example.

[Explanation of symbols]

10 Microphone 11 Voice Information Transmission Unit 12 Information Reception Unit 13 Voice Output Unit 14 Voice Information Reception Unit 15 Database Registration Unit 16 Attribute Extraction Unit 17 Distribution Information Database 18 Database Search Unit 19 Information Distribution Unit 20 Provided Voice Information Storage Unit 21 Voice Recognition Unit 22 Semantic analysis unit 23 Attribute discrimination unit 24 Voice information temporary storage unit 25 Database registration unit 26 Received voice information storage unit 27 Database search unit 28 Preference attribute storage unit 29 Speaker verification unit 30 Speaker information search unit 31 Speaker information storage unit 32 reception voice information storage unit 33 reception voice information storage unit 34 favorable rating determination unit 35 speaker information registration units 100a to 100c information communication terminals 101a, 102a, 103a, 104a information communication terminal 200 Internet 300, 301, 302 center device

[Procedure amendment]

[Submission date] January 20, 2003 (2003.1.2
0)

[Procedure Amendment 1]

[Document name to be amended] Statement

[Correction target item name] 0006

[Correction method] Change

[Correction content]

FIG. 24 shows the overall configuration of the information distribution system. The information distribution system has a plurality (here, three) operated by a plurality of users (here, users A to C).
Information communication terminals 912a to 912c, the Internet 913, and a center device 914. The center device 914 and each of the information communication terminals 912a to 912c are communicably connected via the Internet 913.

[Procedure Amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0086

[Correction method] Change

[Correction content]

Next, the operation of registering the voice data received from the center device 302 in the received voice information storage unit 32 will be described with reference to the flowchart shown in FIG. The information communication terminal 103a receives the voice data from the center apparatus 302 in step S50. In step S51, speaker verification processing is performed based on the received voice data, and a speaker ID corresponding to this voice data is obtained. In step S 52, by referring to the speaker information stored in advance in the speaker information storage unit 31, and acquires the priority information set for speaker ID obtained by speaker verification. In step S53, the received voice data is stored in the received voice information storage unit 32 in association with the priority information and the attribute.

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) G06F 17/30 170 G06F 17/30 210C 210 230Z 230 H04M 11/06 G10L 15/00 G10L 3/00 571T 15 / 08 551A 15/10 531W 17/00 545A H04M 11/06 (72) Inventor Kazuyo Ota 22-6 Motomachi, Kadoma City, Osaka Prefecture Matsushita AVC Multi Media Soft Co., Ltd. (72) Inventor Koji Yamamoto Osaka Prefecture 1006 Kadoma, Kadoma-shi, Matsushita Electric Industrial Co., Ltd. (72) Inventor, Hiroyuki Miyazaki 1006 Kadoma, Kadoma-shi, Osaka F-term, Matsushita Electric Industrial Co., Ltd. (reference) 5B075 KK07 ND14 ND23 NK02 NK13 NK24 NK32 PQ04 UU40 5D015 A KK02 LL12 5K101 KK05 KK18 KK19 LL00 MM07 NN07 NN08 NN15 NN36 NN37

Claims (21)

[Claims] 1. An information distribution system for distributing various distribution information from a center device to a plurality of information communication terminals through a communication line, and a voice input means for converting a voice uttered by a user into voice data. Attribute extraction means for extracting the attribute of the voice data from the voice data converted by the voice input means, and storage for storing the voice data converted by the voice input means in association with the attribute extracted by the attribute extraction means Means, a search means for searching voice data having a desired attribute from a plurality of voice data already stored in the storage means, and the voice data searched by the search means to a predetermined information communication terminal. An information distribution unit for distributing, and a sound output unit for converting the sound data distributed by the information distribution unit into sound and outputting the sound. The audio input means and the audio output means are provided in each of the plurality of information communication terminals, and the storage means,
The information delivery system, wherein the search means and the information delivery means are provided in the center device. 2. The attribute extraction means, a voice recognition means for converting voice data into text data, a semantic analysis means for extracting a meaningful keyword from the text data, and an attribute of the voice data based on the keyword. The information distribution system according to claim 1, further comprising an attribute determination unit that determines 3. The search unit searches for voice data having an attribute corresponding to the attribute of the voice data converted by the voice input unit, from a plurality of voice data already stored in the storage unit, The information distribution unit according to claim 1, wherein the information distribution unit distributes the voice data searched by the search unit to an information communication terminal which is a transmission source of the voice data converted by the voice input unit. system. 4. The center device further includes a preference attribute storage unit that stores the attribute extracted by the attribute extraction unit in association with the user as a user preference attribute, and the search unit includes the preference item. By referring to the preference attribute of the user stored by the attribute storage means, the voice data of the attribute corresponding to the preference attribute of the user of the information distribution destination is searched from the plurality of voice data already stored by the storage means. The information distribution system according to claim 1, wherein 5. The attribute extraction means is provided in at least each of the plurality of information communication terminals, and the information delivery means delivers the voice data searched by the search means together with the attribute of the voice data, The information communication terminal includes a reception voice information storage unit that stores the voice data distributed by the information distribution unit in association with an attribute, and a plurality of voice data stored by the reception voice information storage unit. Further comprising a terminal side search means for searching voice data of an attribute corresponding to the attribute extracted by the attribute extraction means, wherein the voice output means converts the voice data searched by the terminal side search means into voice. The information distribution system according to claim 4, wherein the information distribution system outputs the information. 6. The information communication terminals each temporarily store the voice data converted by the voice input means while the communication line is not connected, until the communication line is connected. The information distribution system according to claim 5, further comprising voice information temporary storage means. 7. An information distribution system for distributing various distribution information from a center device to a plurality of information communication terminals through a communication line, and a voice input means for converting a voice uttered by a user into voice data, Storage means for storing the voice data converted by the voice input means, information delivery means for delivering the voice data stored by the storage means to a predetermined information communication terminal, and voice data delivered by the information delivery means And a speaker information storage unit for storing priority information indicating the priority of the voice data of the speaker for each speaker, A voice output means for converting the delivered voice data into a voice and outputting the voice, the voice input means, the speaker verification means, the speaker information storage means and the voice Force means is provided in each of the plurality of information communication terminals, the storage means and the information delivery means are provided in the center device, and the delivered voice data is stored by the speaker information storage means. The information delivery system is characterized in that, based on the priority information corresponding to the speaker of the voice data, the voice output means converts and outputs the voice. 8. The information communication terminals each determine an impression of a user holding a speaker of a voice output by the voice output unit, based on the voice data converted by the voice input unit. The speaker information registration means for registering priority information for the speaker in the speaker information storage means based on the determination result of the sensitivity determination means. Information and communication system. 9. An information communication terminal for receiving various distribution information distributed from a center device through a communication line, wherein a voice input means for converting a voice uttered by a user into voice data, and the voice input means for converting the voice data. Attribute extraction means for extracting an attribute of the voice data from the extracted voice data, and voice for transmitting the voice data converted by the voice input means to the center device together with the attribute of the voice data extracted by the attribute extraction means An information communication terminal comprising: an information transmitting means; and a voice output means for converting voice data distributed from the center device into voice and outputting the voice. 10. The attribute extraction means, a voice recognition means for converting voice data into text data, a semantic analysis means for extracting a meaningful keyword from the text data, and an attribute of the voice data based on the keyword. The information communication terminal according to claim 9, further comprising an attribute determining unit that determines 11. The received voice information storage means for distributing voice data from the center device together with an attribute of the voice data, and storing the voice data distributed from the center device in association with the attribute, the received voice. The plurality of voice data stored by the information storage means, further comprises a terminal side search means for searching voice data of the attribute corresponding to the attribute extracted by the attribute extraction means, the voice output means, 10. The information communication terminal according to claim 9, wherein the voice data searched by the terminal side search means is converted into voice and output. 12. A voice information temporary storage means for temporarily storing the voice data converted by the voice input means while the communication line is not connected, until the communication line is connected. The information communication terminal according to claim 11. 13. An information communication terminal for receiving various kinds of distribution information distributed from a center device through a communication line, wherein voice input means for converting a voice uttered by a user into voice data and distributed from the center device. Speaker verification means for matching the speaker of the voice data to identify the speaker of the voice data, and speaker information storage for storing the priority information indicating the priority of the voice data of the speaker for each speaker. Means and a voice output means for converting the delivered voice data into voice and outputting the voice, the delivered voice data being stored by the speaker information storage means, the speaker of the voice data. An information communication terminal, which is converted into a voice and output by the voice output means based on the priority information corresponding to. 14. A positivity determination means for determining, based on the voice data converted by the voice input means, an impression that the user has for the speaker of the voice output by the voice output means, 14. The information communication terminal according to claim 13, further comprising speaker information registration means for registering priority information for the speaker in the speaker information storage means based on a determination result by the positivity determination means. 15. A center device for distributing various kinds of distribution information to a plurality of information communication terminals through communication lines, and voice information receiving means for receiving voice data from the information communication terminals, and the received voice. Storage means for storing data in association with the attribute of the voice data; search means for searching voice data having a desired attribute from a plurality of voice data already stored by the storage means; An information distribution means for distributing the voice data searched by the information communication terminal to a predetermined information communication terminal. 16. The apparatus further comprises attribute extracting means for extracting an attribute of the voice data from the voice data received by the voice information receiving means, wherein the storing means extracts the received voice data by the attribute extracting means. 16. The center device according to claim 15, wherein the center device stores the associated attribute. 17. The attribute extraction unit, a voice recognition unit that converts voice data into text data, a semantic analysis unit that extracts a meaningful keyword from the text data, and an attribute of the voice data based on the keyword. 16. The center device according to claim 15, further comprising attribute determining means for determining. 18. The search means searches for voice data having an attribute corresponding to the attribute of the voice data received by the voice information receiving means from among the plurality of voice data already stored by the storage means, 16. The information distribution unit distributes the voice data searched by the search unit to an information communication terminal which is a transmission source of the voice data received by the voice information receiving unit. Center device. 19. A preference attribute storage unit for storing the attribute of the voice data received by the voice information receiving unit as a preference attribute of the user in association with the user, and the search unit includes the preference attribute. By referring to the preference attribute of the user stored by the storage means, the voice data of the attribute corresponding to the preference attribute of the user of the information distribution destination is searched from the plurality of voice data already stored by the storage means. 16. The center device according to claim 15, wherein: 20. An information distribution method for distributing various distribution information from a center device to a plurality of information communication terminals through a communication line, comprising a voice input step of converting a voice uttered by a user into voice data. An attribute extraction step of extracting an attribute of the voice data from the voice data converted by the voice input step, and a storage storing the voice data converted by the voice input step in association with the attribute extracted by the attribute extraction step. A search step of searching for voice data having a desired attribute from the plurality of voice data already stored by the storing step; and the voice data searched by the search step to a predetermined information communication terminal. Information distribution step for distribution, and sound distributed by the information distribution step A voice output step of converting voice data into voice and outputting the voice, the voice input step and the voice output step,
Executed in each of the plurality of information communication terminals,
The information distributing method, wherein the storing step, the searching step and the information distributing step are executed in the center device. 21. An information distribution method for distributing various distribution information from a center device to a plurality of information communication terminals through a communication line, wherein a voice input step of converting a voice uttered by a user into voice data, A storage step of storing the voice data converted by the voice input step, an information delivery step of delivering the voice data stored by the storage step to a predetermined information communication terminal, and a voice data delivered by the information delivery step. A speaker verification step of identifying the speaker of the voice data by comparing the speakers of, and a speaker information storage step of storing priority information indicating the priority of the voice data of the speaker for each speaker. A voice output step of converting the delivered voice data into voice and outputting the voice, the voice input step, the speaker verification step The speaker information storage step and the voice output step are executed in each of the plurality of information communication terminals, the storage step and the information distribution step are executed in the center device, and the distributed voice is transmitted. The data is converted into voice by the voice output step and output based on the priority information corresponding to the speaker of the voice data stored in the speaker information storage step, and output. Delivery method.